Arm Flexes An Incredible 250 Billion Chip Cores Shipped In 40 Years
You may not realize it, but chances are high that one of your devices is rocking an Arm-based chip of some kind. And probably multiple devices. Arm's IP has been used to design an enormous range of system-on-chips (SoCs) that power many Android smartphones, tablets, routers, smart TVs, and so much more. Given that's the case, it's not surprising to see Arm celebrate having shipped a staggering "250 billion Arm cores" to date.
Arm revealed the eye-popping stat in a blog post celebrating the 40th anniversary of the Arm architecture. And while not surprising, it's certainly impressive and underscores the demand for efficient, low power silicon.
"What began as an ambitious project in a small corner of Cambridge, U.K., has grown into the world’s most widely adopted computing architecture, now powering billions of devices – from sensors, smartphones and laptops to vehicles, datacenters and beyond," Arm states in a blog post.
The post goes into Arm's long and storied history, starting with the introduction of the ARM1 chip back in 1985 (Arm's origin story actually dates back even further, if including a prelude to the ARM1). Developed by Sophie Wilson and Steve Furber, the 32-bit chip featured 25,000 transistors—laughably low by today's standards, but cutting-edge for the time—on three-micron technology. It also had just 6,000 gates and no cache.
To put it into perspective, a current-generation Armv9 chip (released in 2021) built on a 40nm nodes sports 100 million gates, along with graphics acceleration and multiple processing cores.
Arm says it was a project born out of necessity that quickly became a new architecture philosophy, one centered on RISC (Reduced Instruction Set Computing). The benefit of RISC at the time was that its comparatively simplified instruction set (versus x86) paved the way to streamline faster chips.
"The internal slogan was 'MIPS for the Masses' (Millions of Instructions Per Second). This mindset separated the Arm architecture from other processor designs at the time that were focused on high-end workstations or mainframes. The Arm architecture was thinking power-efficient, fast, and scalable," Arm explains.
Arm's blog post gets into some of the early challenges the company faced, and how a joint venture between Acorn Computers (a startup with roots in Cambridge, MA), Apple, and VLSI Technology) helped springboard Arm into the behemoth it is today. How much of a behemoth?
In its most recent earnings report, Arm disclosed it raked in $983 million during the third quarter of its fiscal 2025, with net income (profit, basically) reaching $252 million. Arm has grown so big over the years that NVIDIA was unable to clear regulatory hurdles to acquire the company and had to call off a $40 billion deal, resulting in a non-refundable $1.25 billion 'breakup fee' paid out to Softbank.
Arm's celebratory blog post isn't long, but definitely worth a read.
